PROJECTSUMMARY The autism spectrum disorders (ASDs) affect 1% of the world?s population. The syndromes in this diverse family of disorders share three core features: impaired social interactions, communication deficits, and repetitive behaviors. In the United States, 1 in 68 children are now diagnosed with ASDs, a drastic increase over the last few decades. Despite the perceived ?epidemic? of ASDs, there are few effective treatments. In roughly 30% of patients, ASD is associated with epilepsy that is often also refractory to available treatments. Thus,thereisanurgentneedtodevelopbettertreatmentsforthesechallengingconditions. The core symptoms of ASDs are common in Dravet syndrome, an intractable epilepsy with onset in early childhood.Werecentlydemonstratedthatgeneticreductionoftau,amicrotubule-associatedproteinimplicated in neurodegenerative disorders, prevents or markedly reduces epileptic seizures, cognitive deficits, and premature mortality in a model of Dravet syndrome (Scn1aRX/+ mice). More surprisingly, we found that tau reduction ameliorated social impairments, communication deficits, and repetitive behaviors in these mice. Geneticreductionoftaualsoamelioratedsimilarcommunicationdeficitsandrepetitivebehaviorsinaseparate model of ASD (Cntnap2?/? mice). Encouragingly, genetic tau reduction was well tolerated throughout the lifespan, tau knockdown initiated in adulthood also did not cause obvious adverse effects, and complete ablationwasnotnecessary,asevenpartialtaureductionprovidedsubstantialbenefit.Thesedataledtoour central hypothesis that tau reduction counteracts ASD pathogenesis and may be developed into an effectivetreatmentforseveraloftheseconditions. However, several key issues must be resolved before this strategy is ready for clinical development. To determinewhetherASDsubtypesthatdonotincludeepilepsymaybenefitfromsuchatreatmentapproach,we will examine whether genetic ablation of tau prevents or reduces autism-like behaviors in a third independent mousemodelofgeneticallydeterminedASDthatdoesnotdevelopepilepsy,Shank3B?/?mice.Inaddition,an idealASDtherapywouldbeeffectiveevenifitwasadministeredaftersymptomsbecomeapparent.Totestthis possibility,wewillknockdowncerebraltauexpressioninScn1aRX/+micewithantisenseoligonucleotidesafter autism-likebehaviorshavebecomemanifest.Finally,wewilltesthypothesesaboutthemolecular,cellular,and circuitmechanismsbywhichtaureductioncounteractsthecoresymptomsofASD.Inthelongrun,ouraimis toenabletaureductiontobedevelopedintoatreatmentformultipleASDs.Bydeterminingtheconsequences oftaureductiononmolecularregulatorsofdevelopment,circuitconnectivity,andneuronalproperties,wemay also identify additional entry points for therapeutic intervention, to the benefit of patients affected by ASDs or otherdevastatingdiseasesassociatedwithtau-dependentneuralnetworkdysfunction.